#if !defined(TORCH_STABLE_ONLY) && !defined(TORCH_TARGET_VERSION)
#pragma once

// @generated by torchgen/gen.py from Function.h

#include <ATen/Context.h>
#include <ATen/DeviceGuard.h>
#include <ATen/TensorUtils.h>
#include <ATen/TracerMode.h>
#include <ATen/core/Generator.h>
#include <ATen/core/Reduction.h>
#include <ATen/core/Tensor.h>
#include <c10/core/Scalar.h>
#include <c10/core/Storage.h>
#include <c10/core/TensorOptions.h>
#include <c10/util/Deprecated.h>
#include <optional>
#include <string_view>



#include <ATen/ops/_scaled_mm_v2_ops.h>

namespace at {


// aten::_scaled_mm_v2(Tensor self, Tensor mat2, Tensor[] scale_a, int[] recipe_a, int[] swizzle_a, Tensor[] scale_b, int[] recipe_b, int[] swizzle_b, Tensor? bias, ScalarType? out_dtype, int[] contraction_dim=[], bool use_fast_accum=False) -> Tensor
inline at::Tensor _scaled_mm_v2(const at::Tensor & self, const at::Tensor & mat2, at::TensorList scale_a, at::IntArrayRef recipe_a, at::IntArrayRef swizzle_a, at::TensorList scale_b, at::IntArrayRef recipe_b, at::IntArrayRef swizzle_b, const ::std::optional<at::Tensor> & bias, ::std::optional<at::ScalarType> out_dtype, at::IntArrayRef contraction_dim={}, bool use_fast_accum=false) {
    return at::_ops::_scaled_mm_v2::call(self, mat2, scale_a, recipe_a, swizzle_a, scale_b, recipe_b, swizzle_b, bias, out_dtype, contraction_dim, use_fast_accum);
}

// aten::_scaled_mm_v2.out(Tensor self, Tensor mat2, Tensor[] scale_a, int[] recipe_a, int[] swizzle_a, Tensor[] scale_b, int[] recipe_b, int[] swizzle_b, Tensor? bias, ScalarType? out_dtype, int[] contraction_dim=[], bool use_fast_accum=False, *, Tensor(a!) out) -> Tensor(a!)
inline at::Tensor & _scaled_mm_v2_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & mat2, at::TensorList scale_a, at::IntArrayRef recipe_a, at::IntArrayRef swizzle_a, at::TensorList scale_b, at::IntArrayRef recipe_b, at::IntArrayRef swizzle_b, const ::std::optional<at::Tensor> & bias, ::std::optional<at::ScalarType> out_dtype, at::IntArrayRef contraction_dim={}, bool use_fast_accum=false) {
    return at::_ops::_scaled_mm_v2_out::call(self, mat2, scale_a, recipe_a, swizzle_a, scale_b, recipe_b, swizzle_b, bias, out_dtype, contraction_dim, use_fast_accum, out);
}
// aten::_scaled_mm_v2.out(Tensor self, Tensor mat2, Tensor[] scale_a, int[] recipe_a, int[] swizzle_a, Tensor[] scale_b, int[] recipe_b, int[] swizzle_b, Tensor? bias, ScalarType? out_dtype, int[] contraction_dim=[], bool use_fast_accum=False, *, Tensor(a!) out) -> Tensor(a!)
inline at::Tensor & _scaled_mm_v2_outf(const at::Tensor & self, const at::Tensor & mat2, at::TensorList scale_a, at::IntArrayRef recipe_a, at::IntArrayRef swizzle_a, at::TensorList scale_b, at::IntArrayRef recipe_b, at::IntArrayRef swizzle_b, const ::std::optional<at::Tensor> & bias, ::std::optional<at::ScalarType> out_dtype, at::IntArrayRef contraction_dim, bool use_fast_accum, at::Tensor & out) {
    return at::_ops::_scaled_mm_v2_out::call(self, mat2, scale_a, recipe_a, swizzle_a, scale_b, recipe_b, swizzle_b, bias, out_dtype, contraction_dim, use_fast_accum, out);
}

}

#else
#error "This file should not be included when either TORCH_STABLE_ONLY or TORCH_TARGET_VERSION is defined."
#endif  // !defined(TORCH_STABLE_ONLY) && !defined(TORCH_TARGET_VERSION)